OS/2 Past, Present, Future

Bradley Wardell Stardock Systems, Inc.  

December 1997

Copyright notice:
This material is protected under international copyright. If you wish to reprint this article in part or in full, please contact bwardell@stardock.com for permission. You may also visit http://www.stardock.com to see this article and related articles.

Disclaimer:

I work for an OS/2 software company called Stardock Systems, Inc. We develop and publish
32bit OS/2 software for both corporate OS/2 sites as well as individual OS/2 users. Our
products range from complete desktop environments for securing and standardizing corporate
desktops (Object Desktop Professional), to utilities to allow corporations and "power users" to
recover from system hangs (Process Commander) to consumer entertainment products such as
Galactic Civilizations, Trials of Battle, and most recently Links OS/2.

Note: Parts I and II are available at http://www.stardock.com in the "news" section.
   

Part III: OS/2 Future

Our Story So Far...

In the Summer of 1991, IBM made the decision to bring "The Workplace Shell" that had been largely created for IBM’s failed Office Vision project to OS/2. It was and still is marvelous technology that allows people to work the way they want to instead of the way the OS dictates. But doing so meant a delay in the release of OS/2 from a Fall 1991 to a near third quarter 1992 general availability which gave Microsoft enough time to release Windows 3.1 – making OS/2’s Windows support outdated and Microsoft the time to build Windows 95/NT.

Even still, while that move may have ultimately prevented OS/2 from achieving overwhelming dominance, OS/2 was on the path to being a major niche (i.e. 20% of the PC market) platform when IBM PSP division (the one that makes OS/2) got caught up in their early success and decided to jump onto the PowerPC band wagon with both feet. After 2 years and millions of dollars, the PowerPC didn’t cut the mustard and Workplace OS (OS/2 for the PowerPC) was a much tougher task to create than originally anticipated. As a result, the entire thing got scrapped and OS/2 for Intel lost 2 years of development and lost the faith of the executive management of IBM.

In the new reality, IBM created OS/2 Warp 4, a relatively modest upgrade that was supposed to be both a JAVA business client as well as an OS that you could literally talk to. Without enough budget thrown at either target, it didn’t succeed as either one and continued to be used as an industrial strength client for Fortune 200 companies while the SOHO and consumer market began to migrate to Windows 95 and NT 4.0 which unfortunately came out around the same time as OS/2 Warp 4.

Because of IBM’s considerable overhead, they can’t really actively push into a market unless they have overwhelming dominance. OS/2 was really the first time that IBM had really invested to get into a market that already had entrenched mature competitors.

So what’s OS/2’s future?

As 1997 nears its end it seems like OS/2 is slowly being taken over by third parties to greater and greater extents. Ultimately, the only thing IBM really needs to do is provide device driver support and keep their foot out of their mouth long enough for OS/2 to recover from IBM’s mis-steps.

We’ve already seen the beginning of this: Last winter, Esther Schindler, a respected PC journalist, organized an OS/2 Developer’s trade show in Arizona. This Fall, the Felix Cruz inspired Warpstock trade show for end users will be held in Southern California where most of the leading OS/2 vendors will also be to mingle with OS/2 consumers as well as corporate buyers.

But before we get ahead of ourselves, we need to look hard at the reality of OS/2...

It’s not dead but it’s unconscious...

OS/2 users need to know where to set their demands and expectations. I won’t pretend to speak for the other OS/2 ISVs but one thing unique about the OS/2 market is that the ISVs get along very well. The different software companies talk to each other on a regular basis. So let me put it like this, right now, the OS/2 ISV (the companies that make OS/2 software) are in a crisis. None but the smallest (i.e. people who write OS/2 software part time) are going to survive long term as OS/2-only ISVs. They must go cross-platform or into consulting or die. As it stands, there’s just not enough revenue to be made making OS/2-only shrink wrapped products for a company of significant size to thrive.

Let that sink in for a minute.

So what you’re going to see is a lot more cross platform products and some products that are not on OS/2 at all from traditionally OS/2-only companies. It’s not a matter of getting "greedy" it is a matter of survival. If OS/2 ISVs stay OS/2 only without finding some other source of revenue they will go out of business – period. And in this time of transition, it will be particularly important that OS/2 users are supportive of OS/2 ISVs that release cross platform products (and I don’t speak just of Stardock, as the months progress, you’ll see most of the OS/2 ISVs coming out with cross-platform products). So if you were planning to buy or register that OS/2 application or utility, now would be a good time.

The first problem is that OS/2 ISVs face presently is the lack of channels to announce products in. There is no print OS/2 publication in the United States and in Europe there is only OS/2 Inside and little coverage elsewhere. In July, Stardock released OS/2 Essentials Version 2, this is a product that every OS/2 user should want to have. Unfortunately, how does an OS/2 software company spread the word? The Internet can only reach so far.

The second problem is that the OS/2 market isn’t really growing anymore so at some point you’ll saturate the market. As incredible as Object Desktop is, at some point we’ll have sold it to everyone that is going to buy it. How many people reading this don’t already have Object Desktop? This means that a clock is ticking to either expand your market or keep upgrading your product in the hopes that you get nearly a 100% upgrade rate which is still a diminishing return.

Many vocal OS/2 users on the Internet proudly proclaim that OS/2 software has to compete for their buying dollars with DOS and Windows programs since OS/2 can run them. This is true, but it works both ways, OS/2 users have to compete for the development dollars of OS/2 ISVs with DOS/Windows and other platforms too. So while OS/2 software companies have to make it worth the OS/2 user’s while to buy their software, the OS/2 ISVs need something in return and right now (2nd half1997) it’s not happening. The biggest reason Stardock has remained relatively strong longer than other ISVs is because we have continued to release new products (Links golf for OS/2 in May, PMINews in June, OS/2 Essentials 2 in July, BUGS in August, Entrepreneur this October). But obviously we can’t and won’t be able to release a new OS/2 product virtually every month especially when there are not enough channels to spread the word on your product.

So what is going to happen and what can be done? Well, from years of talking to thousands of OS/2 users in my travels I’ve come to notice the different types of people that use OS/2 versus Windows versus Linux.

Windows users are largely end users that don’t like to write software but will happily buy commercial software if it is from a leading brand.

Linux users are techies that like to write freeware but don’t like to buy software but there’s always plenty of software because the user base is always writing little freeware programs.

OS/2 users are largely end users that don’t like to write software but also don’t like to buy software preferring to look for freeware DOS, Windows, and sometimes OS/2 solutions to fill their needs.

Obviously these are generalisms and plenty of OS/2 users like to buy software or there would be no commercial or shareware OS/2 software vendors. But the silent majority of OS/2 users I’ve met use mostly Windows 3.1 programs, maybe a copy of Word 2.0 for Windows that they got from a friend and a bunch of expired beta software / drivers they’ve pulled from BBSes and the Internet.

It’s not that OS/2 users are cheap, it’s that they don’t (as a general rule) value their time to the point that they’re willing to pay for a program that is ten times as good or efficient. Any time I’ve seen a post complaining that Process Commander’s $69.95 price tag is too high I know that the user doesn’t value the time they spend having to reboot because of a system hang or having to stand by and wait for OS/2 to shutdown to reboot instead of letting it auto-reboot the machine for them.

But the same is true of Linux users as well but to a more extreme extent. They’ll just go out and grab GNU C++ and program something that solves a problem they have! But there’s not much commercial or shareware software for Linux as a result.

But this is one of the problems that OS/2 users have to face for OS/2 to have a viable future. They’ll need to either learn to develop their own tools like Linux users or buy (or register) OS/2 products. Let me stress that many OS/2 users do buy (or develop) their computing solutions, it’s just that the majority don’t. An entire market of end users can’t realistically keep up to date if they entirely rely on freeware.

Many OS/2 users don’t realize how much of a difference each user makes, especially right now. Every sale won or lost in the OS/2 market makes a big difference to the software vendor. To put it in perspective, if you sell 30 copies or more a month at Indelible Blue, you’ll be on the Top 10 best seller's list. So imagine selling only a few dozen products per month and having to pay employees, rent, inurance, phone, advertising, etc.? My fear is that unless something happens, all the OS/2 ISVs that are more than a couple of people will be essentially gone by the end of the year unless they’re able to find some success in the Windows platform or migrate to consulting or there’s some sort of increase in OS/2 user purchases.

One of my favorite users on the newsgroups has a tagline that says: "Why do I use OS/2? Because I like the choice!" This is a great motto and one that people who intend to stay with OS/2 should take to heart. A world without OS/2 will be stifled from the PC OS point of view. When Microsoft had no competition, DOS stagnated. What do you think they’ll do if their last major hurdle to monopoly is gone?

So next time you’re manipulating a bunch of graphics files, before you go and fool with a bunch of freeware programs, go and buy PMView which will probably do the job twice as fast and much easier. Or next time you’re fooling with a freeware INI editor that may or may not fry your OS/2 system, check out UniMaint. And yes, next time you want an "OS/2 Equivalent" of Winzip (which is a $30 program on Windows), pick up Object Desktop which, among many things, integrates ZIP files into the OS like a folder.

My mantra on why I use OS/2 is this: "I use OS/2 because it makes me more productive!" Time is money and if enough OS/2 users can come to value their time more, more OS/2 software will be sold.

Why should companies develop for OS/2?

Some companies when they don’t see enough sales on OS/2 begin to get the grass is greener on the other side of the fence syndrome. Their theory is that the Windows world is paved with gold and if they make a Windows program it’ll make them millions of dollars. And if you’ve read this far, you might get the mistaken impression that supporting OS/2 is a waste of time – not true!

The Windows market is bigger – much bigger. But it has a lot more competition. Describe didn’t sell huge amounts of units for OS/2. But it certainly sold a lot more than it would have if it had started out on Windows. Many markets on Windows are closed at this point. Mesa/2 is a fine spread sheet but it is unlikely in the Windows world that it would sell more than a 1000 units. Again, that market is closed.

Imagine if Galactic Civilizations had been written for DOS instead of OS/2 back those years ago? How could a tiny company (back then) like Stardock compete with the millions of dollars games get thrown at them in marketing? Even today with a budget exceeding a million dollars, our next generation strategy game, Entrepreneur (on OS/2 and Windows) has had to fight and scrap to get preview coverage from the mainstream media. But because we have an OS/2 version, we have an advantage both in terms of sales and in terms of word of mouth over a Windows-only game. And word of mouth sells more copies of a product than anything else. Users who get the Object Desktop for Windows (beta) in a few weeks will have probably heard of it because of the OS/2 version (which will also be enhanced in the future).

My point is that the Windows market is very crowded and Windows users like to buy the #1 product in a given market segment. Being the second best widget on Windows puts you in a very poor sales category.. This is where OS/2 development comes into play. I saw Usenet post today that the demo version of Entrepreneur had already made up to being one of the most popular downloads on the Internet Top 40 demos. Anyone here doubt that the amazingly fast climb up the charts is because of OS/2 users? By developing for OS/2 as well as Windows, companies get the support of an enthusiastic user base.

So to summarize to this point, OS/2 users need to value their time and be aware that every purchase they make counts to OS/2 software vendors – especially at this critical time. OS/2 software vendors need to be wary of the myth of the streets of gold in the Windows world – they’re not. Stay with OS/2 at least as a cross platform choice.

The Future We Can Build

This next part will talk about things that are possible to do with OS/2 from a third party standpoint. Don’t take these as product announcements from any particular company. They are just ideas of things that I know could realistically be done with OS/2.

The first thing that needs to be done is to rebuild the marketing channels for OS/2. Os2.announce isn’t sufficient as the primary "marketing" channel. To reach OS/2 users we need to know where they are. We also have to have a realistic idea of how many OS/2 users that can potentially buy products there are. This means setting up a taskforce to eliminate the "14 million" OS/2 user myth. Are there 300,000 active OS/2 users or 3,000,000? The number is somewhere in between. We would also have to clearly define what an active OS/2 user is. Someone using OS/2 to ring in sales as some warehouse which has no third party software it probably not really an active user since they just use it to type into a DB2 database all day. The next step is to find out how active users hear about news. Are they all on the Internet and if so, how regularly? How many of them read print magazines?

One thing we, as a community, can do is to make sure that Internet News is easy to come by. Much easier than it has been. Right now, once it scrolls from the newsgroups, it’s gone. And OS/2 e-Zine isn’t really designed to be a news center.

Recently, there has been a movement to create a thing called "Warpcast" which will send news to you. This is a good start but not quite enough. What we need is a free third party application that when run will go to a central server and let users read about all the latest OS/2 news or a website. A web browser link to a page controlled by a totally unbiased third party that lists OS/2 related news events as the occur and it set up in a searchable easy to use resource. Then, we get every single OS/2 freeware, shareware, and commercial vendor to link to the site and have their program install with a link to this site so that everytime you get an OS/2 product, you get a link to this site. Eventually, everyone that is even remotely active in the OS/2 world would have access to the latest news and events.

Of course, that only covers the on-line portion of the OS/2 market. How do we reach the non-connected OS/2 user as well as the corporate types which is where the "big" money is? Traditionally, Stardock has gone the route of having an active sales force that calls known OS/2 customers. Unfortunately, this is no longer an even remotely cost effective route to take. What we need are print magazines that cover OS/2. Stardock Magazine goes out to about 32,000 OS/2 readers every quarter but hopefully that’s only a tiny fraction of the active OS/2 user population (you can subscribe by emailing kwilas@stardock.net (with a subject of SUBSCRIBE). The solution? Get OS/2 Usergroups more involved. Right now they are relatively isolated. What we need is a world-wide OS/2 user group organization that keeps track of all the active OS/2 user groups and their stats. Each OS/2 user group would try to find out what companies or people in their areas use OS/2 and invite them to join their user group. These user groups could also pool resources to try to be at various trade shows to make sure OS/2 has some sort of presence there outside whatever wacky strategy IBM has cooked up for that week. People would then learn the relative strengths of OS/2.

Between these two mechanisms, the OS/2 community would have a way to educate users about OS/2 and keep existing OS/2 users informed on the latest happenings.

It’s the software, stupid!

OS/2 is going to need Windows 95 support. OS/2 will never get a native CorelDraw caliber application nor will it ever get a top notch desktop publishing package. And while many OS/2 users would stick their head in the ground and say "Product <Insert nice but immature OS/2 app here> is way better than <insert top of its class Windows product here>!", there are certain types of apps that people need and OS/2 will never get them.

Someone needs to provide this. I don’t know who and I don’t really care but somehow, OS/2 must be able to run Win32 programs. Linux can do it, why can’t OS/2?

By the time you read this, Stardock will have probably announced that Object Desktop NT (For Windows 95/NT) is pretty far along (if you’re interested in being in the early experience program, visit http://www.stardock.com, it’ll be $45 to join in but you’ll get access to 2 years up upgrades – only first 5000 users will be allowed in). I mention this because there are things coming up in NT and Win98 that are pretty neat that can be done on OS/2.

For example, the Windows 98 desktop can actually be made up of web pages. That is, instead of having a wall paper you could have a website or a bunch of websites positioned on different parts of your desktop to make your system more productive (or much less depending on what you put on there). I envision an Object Desktop future where we can embed applications and websites directly into the OS/2 desktop. Running an IRC program? Make it part of the desktop. How about an Active JAVA WarpCaster where the latest OS/2 news is part of your desktop?

Another thing that would be nice to see is someone to be able to make Windows programs look like OS/2 programs. It shouldn’t be super hard to do. Stardock has not jumped onto this because of potential support issues but any serious Windows programmer should be able to figure out how to change the controls to look like OS/2 ones.

It should also be possible (though not easy) to hook into the file save message and intercept saved files to allow long file names on FAT where the long file name is kept in the EA. Then, if someone could whip up a CLI replacement that read the WPS EA names instead of the regular DOS file names you would be able to read and write long file names on FAT.

OS/2 also needs NTFS support. Someone has already posted that they could do one and they need to be given support. That doesn’t mean it in the traditional OS/2 user fashion: Flood them with promises to buy and then refuse to when it comes out because it doesn’t have some arbitrary unasked for feature, tell them you would pay $50 or something for such a product and let them put you on a list and when it’s done, buy it!

I could even see the day where there is a non-IBM upgrade to OS/2 created by third parties and put into one shrink wrapped package that works as a true upgrade path for traditional OS/2 users should IBM go the route of stripping down OS/2 and making it some sort of weird JAVA client that will run presentation manager programs, Win 3.1 programs, as well as JAVA programs.

What I’m leading up here to folks is that we need to have an OS/2 Developer’s Network where end users and OS/2 software vendors combine forces to come up with things that OS/2 needs to be an even greater OS and then users would need to pledge to support the vendors that do that before hand. There could be a web page run by an objective third party that is connected to the WarpCaster idea I mentioned earlier that allows people to discuss these things and then pledge to purchase these products if they make them. If something got more than 1000 pledges then it would be listed and OS/2 vendors (shareware, commercial, or even freeware) could go and make such a thing.
 

Summaries

OS/2 is going through a very critical time right now (4th quarter 1997). The market is presently not able to support an OS/2-only ISV that has more than a few employees and thus they must become cross platform to survive. OS/2 users need to accept this and support this since if they go down, there won’t be anymore shrink wrapped or large shareware software for OS/2.

OS/2 users also need to be aware that time is money and not always look for the freeware solution unless it is the best one available. Every sale of an OS/2 product counts and OS/2 products are amongst the highest in quality on any market.

IBM’s path for OS/2 is one of leveraging their position in Fortune 200 companies. They want to turn OS/2 into the niche OS that runs "a bunch of stuff" in your workplace but mainly as a JAVA client. This is pretty incompatible with the majority of active OS/2 users needs.

Therefore, OS/2 users and vendors need to work together to control their own future with OS/2. IBM’s direction will ensure that they keep up with driver support which is the only thing third parties cannot do.

A WarpCaster web page controlled by a non-interested third party that is linked to by every OS/2 ISV and has a link created with the installation of every OS/2 program would help everyone. An international OS/2 user group organization that works closely with OS/2 ISVs would help bring OS/2 to new people as well as keeping those who aren’t on the net regularly up to date with the comings and goings of OS/2.

And finally, an organization that comes up with features and enhancements OS/2 needs to stay state of the art. This organization would then work with the OS/2 ISVs to create these components after enough OS/2 users pledged to support these endeavors. Eventually, if enough new features were created, a third party upgrade to OS/2 could be made (like Power OS version 5 or something) if IBM totally dropped the ball.

OS/2 will never be the dominate OS. Windows 98 will be and eventually Windows NT will be after that. But I could see a strong OS/2 market again if the ISVs can survive the transition and the users get organized. Strong enough to keep Microsoft on their toes and keep OS/2 the OS for the rest of us.

I think by getting organized OS/2 can remain a viable OS and be in better shape than it is today. I also think that if OS/2 users, as a whole, can become more aware of how they affect their own OS/2 computing future that native OS/2 software will continue to flow.

OS/2 has had a truly epic history and with any luck it's just beginning. But unless the current trend is changed, in 10 years we'll be using descendents of Windows NT.